Last Modified on June 03, 2012. Please keep us updated if you adopt this model and make improvements.
Authors: Tiberius Brastaviceanu, Suresh Fernando, Anthony Brasher, Arlin J. Wallace, Roy Zuniga add your name
Decision making for an open enterprise
work in progress...
You want to contribute? Put your name up there and have fun!
This document is open for everyone to contribute.
If you contribute to this doc make sure you respect the Content rules
New additions are marked like this. Last addition was made on June 22, 2011
You may use this text as you please!
Collaboration: don’t delete text/ideas, you can freely improve the style, you can freely reformat the text to explicit its thematic and pragmatic structure, use comments and discussions, chose a different color for editing if you introduce new ideas, be respectful.
Color attribution: Tiberius Brataviceanu, Anthony Brasher, ...
Questions at the bottom of the document..
Please contribute with references.
Table of contents
This reflexion is principally intended for a resource-intensive open, decentralized and self-organizing value network, (call it an open enterprise) designing, producing and distributing material products. It can also apply to other type of open and decentralized projects.
An open enterprise is a value-based network, not a power-based hierarchy. Moreover, it is an environment in which ideas, initiatives, and projects compete with each other, NOT individuals.
Hierarchies have decision makers. The decision making process can take many forms, but in the end, one individual is empowered to cut through the debate, and bears the responsibility of the consequences of his decision. It is assumed that the individual in charge has the capacity to weigh all the costs and benefits, but this is not always the case. Democratic decision making (a majority vote for example) is also used, in which case the responsibility is shared. The democratic process can be broad-based or selective, including only specialists.
An open enterprise is decentralized. It is based on the assumption that contributors have equal potential (which is NOT to say that they all actualize their potential equally). It is an empowering environment, stimulating everyone to passionately contribute with value to the venture. It is a fluid value system in which good initiatives have a greater chance to be implemented, even if they arise on the “margins” of the network.
The open enterprise has a distributed decision making system, it is flat or decentralized in that sense, but it is not flat with respect to its processes. Contributors need to understand their role within the value production and distribution system, and need to subordinate to and coordinate with other members in order to increase the value (market value of the product) in between them, which binds them together. That subordination is voluntary, but it naturally arises if every member understands how value is to be created and redistributed, which forms the set of positive incentives that drive every one's involvement. So a role system is needed.
When it comes to making decisions, every contributor needs to understand not only every other contributor’s role and capacity, but also the reputation. A reputation system is needed. Reputation is a complex. It’s not just a number, or the position of a cursor on a slide-bar. In fact, reputation is strongly related to predictability in context. It is a complex of indices, acquiring their value mostly from the past, but not only, which help us determine the probability that a player of a game will act rationally, to increase his/her reward, and the value/strength of the entire community, which is in part transferred back to the same player as reward. In other words, the reputation of a member of a value network is the probability that this member respects the rules that govern co-creation and transactions within the network, which, if the network is properly designed, should maximize reward for the member, which in turn aligns with what makes the network sustainable (there should be no conflict between individual interest and collective interest!).
There are different types of decisions to be made within an open value network. Some of them refer to incremental additions to a project. Others determine a path among different possible ones, not only binary cases. Decisions over allocation of internal resources should be less common within an open value system, because resources and capacity can always be increased by integrating external structures. (allocation of resources is bottom up)
What are the consequences of NOT having someone invested with the power to cut through and make a decision? Whenever a new initiative arises, some contributors who don’t understand very well the value system can oppose measures, even disregarding the reputation structure. The network can be paralyzed in interminable debates. Obviously politics becomes much more complex within a decentralized network. How can this problem be circumvented to allow the group to focus on constructive activity rather than losing its focus and energy on lengthy debates?
One can suggest a democratic process, but we all know that a vote only reflects the lowest common denominator, and smart decisions can be blocked by an ill-informed majority. Somehow reputation must be taken into consideration (a selective democratic process). But we don’t want to create a meritocracy, or even worse, to encourage elitism. We want to allow brilliant initiatives arising from any part of the network to express themselves. There are cases when a vote is justified, if it is preceded by a wide process of education and analysis. But in the case of decisions regarding incremental additions to the venture, democracy fails miserably, because most of the time decisions are about very technical issues, and it would imply voting on almost every little addition (which would constitute a big distraction for the entire group).
I see one way out.
A good initiative is one that adds value to the venture and, by the same token increases the rewards for members. We need to consider the open and decentralized enterprise as a Darwinian environment for ideas, proposals, initiatives, in which the good ones get traction and the bad ones are neglected/forgotten. Some conditions must be fulfilled in order for this to happen:
A rule of thumb for an open and decentralized enterprise is to grant automatic permission to implement and test new initiatives. It is the responsibility of the initiator to rally support and resources for his/her initiative. If the initiator can do it alone the initiative cannot hurt the group, even if it does not represent a great addition of value.
This process transfers the decision making to every stakeholder and participant. Everyone makes an individual decision according to how benefits and costs are estimated at the individual level. These individual decisions aggregate and influence each other within the open space. The initiative is given a chance to get started and its progressive and continuous implementation further influences other member's decisions. The collective decision making process becomes continuous, fluid, and is constantly coupled to the initiative itself (the object of decision). It the starting initiative produces good results more members will allocate resources for it, which fuels the initiative even more, and so on... That is what we mean by traction. Negative effects are also rapidly felt and propagate through the system, making bad initiatives short lived.
What if everyone starts implementing initiatives, wouldn’t that defocus the group’s activities and stretch thin its resources? Well, if the value system is well understood and if communication flows freely everyone will realize how to get to the rewards in the most effective way. Before members will invest themselves into one initiative or another they will weigh the benefits and the risks themselves. Some members might get it wrong and adopt a bad initiative, but most of them will not. The individual choices of members with higher reputation will have an impact on other members. This is how skills, talent and capacity are included into this process.
Moreover, open enterprises should have two diffuse layers, a very pragmatic core, composed of goal-driven individuals, and an outer layer composed of passionate, idealistic, fun seeking individuals. For a revenue-driven open enterprise we can say that we have a market and a gift economy intertwined. The value system needs to be conceived in such a way to avoid conflict between these two classes of members, participating on the extremes of these two types of economies, within the network. In other words, the revenue-driven member should have no business to put pressure on passion-driven member, or vice versa, because both contributions should represent value added for the network (if the network is built properly).
My advice is to NEVER spend energy killing new initiatives. Let them run and the good ones will get traction and will be implemented. The bad ones will be forgotten. This is NOT a perfect system, whatever that means! But it is one which, in my opinion, solves the decision making problem in a value-based decentralized environment.
From Roy: Federated Decision Making
Peer governance in peer production (P2P fundation)
Suresh: I am a bit confused since I see that you are suggesting both the you have a maximally open and transparent environment that allows all ideas to be presented since the good ideas will rise to the top and get support, as well as suggesting the need for a reputation system. I am not sure why you need both.
Tibi: The reputation mechanism ensures that the opinions and actions of members are perceived differently by other members, according to the reputation indices, which will influence their own decisions. Reputation doesn't grant power, but exerts influence on others.
Suresh: Furthermore, I am not sure what exactly you mean by a reputation system and exactly how it would be implemented in an open environment.. Maybe this the real source of my confusion ;-)
Tibi: BetterMeans has already implemented a (Ben suggests: joint review--not exactly reputation) mechanism. How do you qualify/quantify reputation? I don’t know... there are many models...
Ben: yes this is tricky, as money and politics and conventional hierarchy is a type of reputation system too. Difficult to codify. Perhaps the informal/social one we all sub/consciously operate can and does go a long ways here already. BetterMeans is a nice simple peer-review system for accountability, still based on a non-uniform team.
Ben Brownell: Curious about establishing and evolving the underlying value system. Or is this taken to be predetermined/external/empirical? Thanks for assembling this TB!
Arlin J. Wallace: I am inherently averse to a “merit/reputation” system. I think direct democracy works best. By direct democracy, I mean “voting” for projects/business opportunities for which YOU will implement and sustain. This type of direct democracy works within a framework of a community economic ecosystem. This framework is created by gathering primary economic, social and psychographic data from a particular geographic area. This data would inform a needs and wants analysis, from which, the community, in an individual and collective way, decide what business, social institutions and infrastructure is required to develop and sustain local self-containment. Once the needs analysis is created, all of the required components are qualified (based on feasibility and economic expediency) and enumerated (non-duplicatively). The community members who voted (via the primary data survey) for their respective components become the proprietors of those components. Two issues are resolved using this model; One, by direct democracy, individuals are engaged and empowered to express a concern and to implement its solution and two, a development and implementation framework is created scientifically (based on empirical primary data, as opposed to majority-rules political ideology), thus ensuring the implementation of intelligent initiatives that from the outset, were derived (or arrived at) prerequisites to the sustainability of the community’s economic ecosystem.
[Tibi] - take a look at “liquid democracy”
Definition: A decision is the process through which an individual chooses a subset of actions from a set of all possible conceivable actions in a given circumstance.
Making a decision is making a choice. Decision making presupposes free will, it is incompatible with determinism. There are other variants on the spectrum from free and determined:
Decision making also presupposes that we have control over our own actions among the physically possible actions in a given circumstance. For example, it is physically possible to play with a ball outside, but this might not be so for a prisoner, who doesn’t have total control over his own actions.
The choice can be free of influence. For example, the election process, which is a collective decision (actors are all citizens who vote) that has as scope the entire nation, is not entirely free. The opinions of voters (actors) are heavily influenced/manipulated by propaganda from different sides.
A decision is NOT choosing among different outcomes, because there could be risk involved, i.e. a suspected outcome is not 100% probable.
The choice can be made knowing with certainty the outcome of some or all possible actions, or can be made in absence of certainty. For the second case, the outcome can be negative, positive, or neutral, which will affect the decision making process.
Making a choice also implies some knowledge about the situation.
The choice can be made at random. This happens when the actor(s) have no information on the situation. It can also happen when other decision making mechanisms are absent.
The choice can be educated, after an effort of analysis. Voluntarily biased analysis amounts to manipulation and to an influenced decision. There is a distinction between genuine analysis and understanding and false/deceiving analysis and explanation.
A decision making process has actors (individuals involved in choosing a set of actions among a set of all possible actions) and scope (the group of individuals that are affected by the choice).
Distinction between making us one’s mind, according to a set of criteria and implementation, which can be done in various ways. Some decisions become effective as soon as they are uttered (ex. the decision of a judge).
Distinction between reversible decisions (ex. getting married) and irreversible decisions (ex. commit suicide). Reversibility can be only partial, because decisions have consequences that cannot be reversed, as we cannot reverse time (ex. after a divorce- the reverse decision to getting married, the family estate gets divided).
From Roy: (…) [a] dialog, (...) around scenarios requiring decisions. (...) write up one or two HIGH PRIORITY representative scenarios (...). Be sure to include:
- the Actor -- who is wanting something
- the Action -- what is the ask
- the Business Value -- why is the ask being made
- the Exit Criteria -- how do we know the request/action is complete
For now, I can distinguish two very different types of scenarios related to:
Classical organizations have a fixed architecture. SENSORICA is a living system, which means that its structure is free to change according to environmental pressures. Members have the power to modify the very nature of the organization, including its infrastructure. In order to make the organization stable and efficient some inertia must be added in appropriate proportions. In other words, the possible rate of change must be close to the natural rate at which important, life threatening, change occurs within the environment. If the system is totally free to transform itself it can become unstable, subject to large fluctuations caused by individual urges, mass emotional waves, or by small and temporary/reversible environmental changes. In this case, too much energy would be spent on constantly adjusting, rather than on pursuing individual and collective goals, and on production.
Moreover, structural changes should concern every member of SENSORICA. Therefore, all members are called in to participate in the decision making process. Should-we apply direct democracy, one member, on vote? Should we apply meritocracy? Should we weigh votes according to equity? Lasy democracy or direct democracy? My opinion is to go with direct democracy, one member, one vote, to make sure that change can also emerge on the edges of the network, and to be able to integrate into the process a larger perspective.
Let’s now move to the second point. Decisions that don’t have the power to affect the structure of SENSORICA don’t necessarily concern all members. For example, a member can propose a new feature to the Mosquito sensor. In order to implement this new feature this member needs to build a team within SENSORICA, and to focus some resources on this particular project. For this kind of initiative the decision making doesn't need to constitute a decision event. It can be a continuous process through which other members allocate attention and resources to it, based on the interest it generates for everyone. SENSORICA must contain analysis and evaluation nodes, i.e. teams that specialize in feasibility studies, market analysis, environmental impact, etc. These open and dynamic teams add or subtract merits to initiatives, dynamically, as they are being implemented.
In other words, there is no group decision for implementing such initiatives. This type of decision is not binary at the group level. But every member makes an individual decision to allocate attention and resources to a particular initiative. Because our value system is open and transparent, for every initiative one can see who is contributing and what resources are already allocated to it, influencing one’s decision to participate. If an initiative gains traction and succeeds in shifting a critical mass (with respect to the initiative) of the community’s resources to it, we can say that the community “has decided” that this particular initiative is worth pursuing.